moteur de recherche

Workshop on Discrete Geometry and Mathematical Morphology for Computer Vision (DGMM4CV)
Workshop on Discrete Geometry and Mathematical Morphology for Computer Vision (DGMM4CV)
24-nov.-2016 09:00
Il y a: 215 days


Taipei, Taiwan, 24 November 2016, in conjunction with ACCV 2016.

Invited speakers

Robin Strand (Uppsala University, Sweden): Digital distance functions - recent advances in theory and applications.

Adrian Sheppard (Australian National University, Australia): Analysis of complex material structure from 3D images using discrete Morse theory and persistent homology. 

Punam Kumar Saha (The University of Iowa, USA): Skeletonization and its application to quantitative morphometry.

Call for Papers

Discrete geometry plays a fundamental role in research fields such as image analysis, computer vision, computer graphics, pattern recognition, and shape modeling. This is because all data in the computer are unavoidably discrete. The foundation of discrete geometry comes from the necessity of the treatment of digitized models or images of objects in the 2D or 3D Euclidean space. Mathematical morphology, on the other hand, is a theory and technique for analyzing and processing geometrical structures based on set theory, lattice theory, and topology. Mathematical morphology is most commonly applied to digital images, but it can be employed as well on graphs, surface meshes, solids, and many other spatial structures.

The community of discrete geometry and that of mathematical morphology have closely communicated and mutually exchanged latest research results and ideas to stimulate and widen the communities. However, the computer vision community has less communicated with both the communities in spite that they all are working for digital images. Sharing recent findings in each community with the other communities contributes to advancing the cutting edge of researchesfor image analysis. There are increased demands to exchange latest results and ideas to foster these three communities together, and it is quite timely to bring them together.

Based on the recent development and discovery in 2D and 3D image analysis, the goal of this workshop is to light up and share common digital and discrete methodology in various fields and to open a new direction of computer vision, discrete geometry and mathematical morphology. Successful researchers are expected to submit their latest results and new ideas in discrete and digital geometric methods in image analysis and its related areas.

Main topics are, but are not limited to:

        Geometric descriptors
        Object digitization
        Geometric transformation
        Geometric motion analysis
        Graph-based method
        Markov random field
        Discrete and combinatorial optimization
        Connected operators
        Hierarchical analysis
        Discrete and computational Topology
        Discrete calculus


        Low-level vision, image processing
        Denoising and filtering
        Segmentation and grouping
        Object detection
        Model fitting
        Point cloud processing
        Image registration
        Surface generation
        Motion segmentation
        Video segmentation
        Motion tracking
        Biomedical analysis
        Human action recognition
        Scene labelling and understanding
        Medical image processing

Organization Committee
Jean CoustyYukiko KenmochiAkihiro Sugimoto

<- retour: